Skip to Content
0

Unexpected disconnection from server (11)

Dec 01, 2016 at 12:33 PM

213

avatar image

Hi,

i am facing the below issue

Updating complex table 'ctClassification'...

Field Definitions received Index Definitions received - rebuilding
Unexpected disconnection from server (11)
Connection failed
Ending transmission

Below are our my server and configuration details

SMP 3.0 SP11 WM6.3 application

data in ctclassification table is 240 records

i have maintained below values for time intervals in SMP Cockpit for my application

HTTP Connect Timeout: 2000

HTTP Receive Timeout: 2000

HTTP Resolve Timeout: 2000

HTTP Send Timeout: 2000

Timeout: 2000

Agentry Editor plug-in version i am using is 70.13.1.15

i have not made any changes in standard ctClassification complex table

other user is able to do sync with out any issue connecting to same server and application.

capture.png (9.7 kB)
10 |10000 characters needed characters left characters exceeded
* Please Login or Register to Answer, Follow or Comment.

1 Answer

Mark Pe
Dec 07, 2016 at 05:36 PM
0

Madhukar,

Hi. Error 11 is a special type of error. It means that the system ran out of threads to perform the needed functions. Based on your statement, you have hit the issue on ctClassification area which only has 240 records. This seems to be a small number of records to keep on getting an Error 11.

My suggestion is for you to understand the issue of Error 11:

1) Error 11 is running out of threads count.

Factors that can contribute to this are at no particular order:

A) You are running out of memory (rule of thumb: There is an application cap of memory. I always find this to be somewhere around 25% of available RAM (physical) + you need to set your JVM memory to land somewhere in that range. Example: If my memory analysis (Run command: jconsole PID - where PID is your Process ID of your Java SMP 3.0 process) is landing around 10Gig. I need to make sure as best practice that my minimum heap size and maximum heap size is somewhere in that area + my physical RAM has 10 Gig as 25%. This is one technique - does not guarantee flawless execution but you will be one step closer and have minimum or reduce down time.

B) You need to look at your Agentry logs and check how many threads you are using (ex: during the sync of peak time are the threads stating 700 or 1000 or 1500). You can use the same jconsole PID command to check your peak threads. With this knowledge, adjust the SMP 3.0 cockpit thead counts (from 250 to your peak thread).

C) Verify that the complex table run time is under 2 minutes or even less. If not then it may be holding on to some process that may need to be optimized. Faster running BAPI queries are known to have better performance than slower run time. Example: If you can tweak the queries (by either doing better filters or re-coding it) to have a run time of < 1 minute versus 15 minutes, you will notice improve performance to prevent Error 11.

D) You may review the SAP Support Cheat sheet of all Error 11 issues here:

https://launchpad.support.sap.com/#/notes/2255046

If ever you need to find the hint where all of this cheat sheets are compiled for all of Agentry just Google keywords: "Agentry Troubleshooting Guides" and find the Wiki of the Master list. The direct link is here:

https://wiki.scn.sap.com/wiki/display/SAPMOB/Agentry+Knowledgebase+Notes+Troubleshooting+Guides (we update this regularly).

Regards,

Mark Pe
SAP Platinum Support Engineer

Share
10 |10000 characters needed characters left characters exceeded